Personality and Individual Differences 76 (2015) 123–128
Contents lists available at ScienceDirect
Personality and Individual Differences
journal homepage: www.elsevier.com/locate/paid
Moral disengagement, the dark triad, and unethical consumer attitudes
Vincent Egan a,⇑, Natalie Hughes b, Emma J. Palmer b
a
b
Centre for Family and Forensic Psychology, YANG Fujia Building, University of Nottingham, Wollaton Road, Nottingham NG8 1BB, UK
School of Psychology, University of Leicester, Henry Wellcome Building, Lancaster Road, Leicester LE1 9HN, UK
a r t i c l e
i n f o
Article history:
Received 8 September 2014
Received in revised form 26 November 2014
Accepted 27 November 2014
Available online 22 December 2014
Keywords:
Dark triad
Personality
Psychopathy
Moral disengagement
Narcissism
Machiavellianism
Unethical consumer attitudes
White-collar crime
a b s t r a c t
Purpose: Bandura’s theory of moral disengagement explains how otherwise ethical persons can behave
immorally. We examined whether a trait model of general personality and the ‘‘dark triad’’ underlay
moral disengagement, the relationship these constructs have to unethical consumer attitudes, and
whether moral disengagement provided incremental validity in the prediction of antisocial behaviour.
Methods: Self-report data were obtained from a community sample of 380 adults via an online survey
that administered all measures.
Results: Correlations between unethical consumer attitudes, lower Agreeableness, lower Conscientiousness, higher moral disengagement, higher psychopathy, and higher Machiavellianism were captured by
a single factor. When this broad factor was examined using regression, demographic, personality and
the dark triad traits all predicted moral disengagement, specific influences being age, education, Intellect,
psychopathy, and Machiavellianism. A similar model examining predictors of unethical consumer attitudes again found all blocks contributed to the outcome, with specific influence provided by age, Intellect,
and moral disengagement, the latter showing incremental validity as a predictor of unethical consumer
attitudes.
Conclusions: Moral disengagement is based on low Agreeableness, Machiavellianism and psychopathictype traits, but provides incremental validity in predicting antisocial attitudes to a trait model alone.
Narcissism is neither related to moral disengagement, nor unethical consumer attitudes.
Ó 2014 Elsevier Ltd. All rights reserved.
While people generally know right from wrong, some find it
easier to disengage from their ethical principles than others. This
behaviour is called moral disengagement. A common example of
moral disengagement is consumer dishonesty, which is defined
as: ‘‘the moral principles and standards that guide behavior of individuals or groups as they obtain, use, and dispose of goods and services’’ (Muncy & Vitell, 1992, pp. 298). Apparent ‘‘petty’’ dishonesty
(Egan & Taylor, 2010) harms UK businesses to the cost of at least
£294 million pounds per annum (National Fraud Authority.,
2011). The current study examined general and darker personality
traits underpinning moral disengagement, using unethical consumer attitudes as a specific criminological outcome.
Moral disengagement (Bandura, 1986) provides a specific
model to explain how persons breach their personal ethics. Individuals generally seek consistency in held moral beliefs to avoid
a discrepancy between what they believe to be right, and how they
⇑ Corresponding author.
E-mail addresses: [email protected] (V. Egan), nathughes@
hotmail.co.uk (N. Hughes), [email protected] (E.J. Palmer).
URL: http://www.nottingham.ac.uk/medicine/about/psychiatryandappliedpsychology/people/vincent.egan (V. Egan).
http://dx.doi.org/10.1016/j.paid.2014.11.054
0191-8869/Ó 2014 Elsevier Ltd. All rights reserved.
actually behave. This is because conflicts between inconsistent
behaviours and beliefs classically produce feelings of psychological
discomfort known as cognitive dissonance (Festinger, 1957). If one
can disengage from personal moral standards, it becomes easier to
justify engaging in behaviours normally considered immoral.
Bandura, Barbaranelli, Caprara, and Pastorelli (1996) described
eight mechanisms by which disengagement of morals may occur;
moral justification (justifying a wrongful act as virtuous in terms
of the perceived potential outcome); euphemistic labelling (using
language to conceal guilt by distorting what happened, as when
a thief says they ‘‘found’’ a stolen item); advantageous comparison
(justifying a wrongful act by fallaciously comparing it to another’s
more egregious acts); displacement of responsibility (when an
individual’s wrong-doing is attributed to being under pressure or
orders from another); diffusion of responsibility (where a shared
decision to behave immorally means no individual involved in
the wrongful act believes they are fully culpable for the events
which occur); disregarding or distorting the consequences (ignoring or minimising the outcome of the wrongful action); dehumanisation (rejecting the human qualities of one’s opponent and seeing
them as bestial); and attribution of blame (suggesting the blame
for wrongful action lies in the provocation of the victim who
124
V. Egan et al. / Personality and Individual Differences 76 (2015) 123–128
brought the trouble on themselves). Such moral disengagement is
seen in rationalisations for antisocial and delinquent acts (Bandura,
Caprara, Barbaranelli, Pastorelli, & Regalia, 2001; Shulman,
Cauffman, Piquero, & Fagan, 2011).
Moral disengagement is significantly and positively correlated
with unethical workplace decision making and behaviour, and is
an indirect proxy for risk of white-collar offending (Barsky, 2011;
Christian & Ellis, 2013). For example, Detert, Treviño, and
Sweitzer (2008) found higher-order qualities of personality such
as empathy and moral identity (the degree moral concerns are perceived as part of your own identity; Aquino & Reed, 2002) correlated negatively with an individual’s propensity to morally
disengage, whereas general cynicism and chance locus of control
orientation (the belief that chance determines an individual’s outcome) were positively correlated with the construct. They also
found the relationship between higher-order personality and
unethical decision making mediated by moral disengagement.
Moore, Detert, Treviño, Baker, and Mayer (2012) subsequently
found positive significant associations between moral disengagement, self-reported unethical behaviour, and self-reported decisions to commit fraud. While these results are salutary, few
studies have examined more fundamental influences underpinning
moral disengagement or it’s correlates, for example, the Five-Factor/Big Five Models of personality (FFM/BFM; Goldberg, 1999;
McCrae & John, 1992), or the negative dispositional traits found
within the dark triad (Paulhus & Williams, 2002)
The dark triad (DT) comprises three constructs; psychopathy,
narcissism, and Machiavellianism. Though scores on the DT correlate with traits derived from both FFM and BFM conceptualisations of general personality (Jakobwitz & Egan, 2006; Pailing,
Boon, & Egan, 2014), Agreeableness has the greatest relative
importance for DT prediction (O’Boyle, Donelson, Banks, Story, &
White, in press). Exemplifying O’Boyle et al.’s observations,
Egan, Chan, and Shorter (2014) found Machiavellianism and psychopathy driven by low Agreeableness alone, whereas narcissism
presented a much more complex FFM profile. This finding
reiterated narcissism as the ‘lightest’ dimension of the DT
(Furnham, Richards, & Paulhus, 2013). When measured concurrently with narcissism, psychopathy and Machiavellianism are
better predictors of unethical and antisocial outcomes (KishGephart, Harrison, & Treviño, 2010; Tang, Chen, & Sutarso,
2008). O’Boyle et al. (in press) argue that given Machiavellianism
and psychopathy are highly correlated, and show a similar pattern of correlations in relation to the FFM, these two constructs
may be better regarded as a single psychopathic entity, albeit
one differentially expressed. This view has a precedent in the
work of McHoskey, Worzel, and Szyarto (1998).
There have been a number of specific studies using models from
social psychology examining aspects of moral disengagement.
Neutralisation, false-returning purchased items as ‘‘faulty’’ or
‘‘unwanted’’ (associated with thrill-seeking and less self-consciousness), anticipated guilt following unethical consumer activity, and
the influence of guilt and opportunism on receiving too much
change at a supermarket checkout have all been examined (De
Bock & Van Kenhove, 2011; Harris, 2008; Steenhaut & Van
Kenhove, 2005, 2006). Moore et al.’s (2012) systematic studies into
moral disengagement in occupational settings measured Machiavellianism alongside measures of empathy and perspective-taking,
finding these predictors associated with a greater ability to morally
disengage, their effect was replicated across two studies, while a
fourth study found dispositional guilt negatively correlated with
the propensity to morally disengage. All of these studies touch on
aspects of the relationship between general personality, moral disengagement, the dark triad, and unethical consumer activity, but
none studied the dispositional foundations upon which their findings arguably stand.
Bandura (1986) suggested the cognitive pathways by which
moral disengagement shapes antisocial behaviour are similar to
those generally rationalising interpersonal aggression and delinquent conduct. The current study explores the relationship
between general personality, moral disengagement, the dark triad,
and unethical consumer behaviour, examining two questions.
Firstly, how much is moral disengagement driven by more basic
dispositional traits, in particular low Agreeableness, low Conscientiousness, and the dark triad. Secondly, does the addition of moral
disengagement provide incremental validity to a basic trait model
predicting unethical consumer attitudes in which low Agreeableness, low Conscientiousness and the DT are expected to explain
most of the variance?
2. Method
2.1. Sample and procedure
This study used a within-subjects correlational design, and was
approved by the research ethics committee. G-Power (Faul,
Erdfelder, Lang, & Buchner, 2007) suggesting 108 persons were
needed for a medium effect size at a statistical power level of 0.8
when using a .05 criterion for statistical significance of P < .05.
The study was administered on-line via GoogleDocs and 382 persons participated. Having oversampled, neither type 1 nor type 2
errors were a concern, and our effect size became more important.
To focus on the most important effects, we adopted a significance
criterion of P < .01 or above.
A convenience participants sample were recruited through
social media, the internal email of several businesses, word of
mouth, and a university student participation resource in exchange
for course credits. All persons gave informed consent. Two participants were identified as outliers on the psychopathy and general
personality measures, and were excluded from the data set, giving
a final sample of 380 participants (271 females, 109 males). Over
half the final sample (51.90%) was employed (41.1% full-time,
10.8% part-time). Under a third of the cohort were students
(28.2%), 11.3% were retired, and 1.8% unemployed. The mean age
was 37.99 years (SD = 16.50, range 18–83, median age = 34).
2.2. Measures
In all cases, scales used Likert responding, with items ranged
from 1 (strongly disagree) to 5 (‘‘strongly agree’’), 3 being a neutral
response. Some items in the scales were reverse-keyed; items were
corrected as necessary before analyses.
The International Personality Item Pool: (IPIP-50: Goldberg,
1999). The IPIP is a 50-item measure of the ‘‘Big Five’’ personality
dimensions; extraversion, Agreeableness, Conscientiousness, Emotional Stability and Intellect. Each of the five scales was measured
using 10 items, with a mean reliability of 0.84 for the five scales
(Goldberg, 1999). Greater scores indicated higher levels of the
dimension measured.
The Muncy–Vitell Consumer Ethics Scale: (CES: Vitell & Muncy,
2005). The CES is a 31-item measure with seven subscales, all of
which are reliable and valid. Four of the subscales relate to unethical consumer attitudes; actively benefiting from illegal actions
(ACT; 5 items), passively benefiting (PAS; 6 items), questionable
but legal actions (QUEST; 5 items) and media downloading (DL;
2 items). (The other three CES subscales relate to positive ethical
behaviours; recycling (4 items), behaving in an honest way (4
items), and activities perceived as involving no harm (e.g., recording a programme off the TV rather than buying the DVD; 5 items)).
For the purpose of this study, greater unethical consumer
behaviour was defined by higher scores on the sum of three of